Pagination

8 Jun 2007 - 12:37pm
7 years ago
33 replies
2569 reads
Allison Beckwith
2006

I have read several design patterns for pagination, and they all say
to format page links like so:

previous link, page links, next link

I am wondering if this is really the best way to organize this. I
have always thought that page links and previous/next links should be
kept separate. Like so:

page links (left aligned) previous link, next link (right
aligned)

When I am moving through pages, I an either I am looking for
something in a certain page range or I just want to move through
pages in order. I use the next link the most, so I like it to stay in
the same spot so I don't have to move my mouse and can quickly click
through pages.

I am wondering if anyone else has thoughts on this.

-Allison

--
Allison Beckwith
Co-Founder & Experience Director

PLANET ARGON, LLC
Ruby on Rails Design, Development, Consulting & Hosting

www.planetargon.com

+1 503 445 2457
+1 877 55 ARGON [toll free]
+1 815 642 4068 [fax]

Comments

8 Jun 2007 - 12:56pm
Todd Warfel
2003

We've actually tested a number of different types of pagination
models and have settled on the following based on time, effort, and
satisfaction performance measurements from several rounds of testing
on a variety of different products over the course of the past 6 years:

Default: < 1 2 3 4 5 > of (N) (where N is the last page)

Once the person has moved outside the initial first five: 1 < 6 7 8 9
10 > (N) (where 1 is the first page and N is the last page)

Once the person has reached the end of the list of pages 1 < 12 13 14
(where 1 is the fist page and 14 is the last page and you're on 14 so
there is no next)

What we found was that the general goal and behavior was to:
* get to the first page
* go to the next page
* jump to the last page
* view ahead to see how many pages are left

And we keep them all together in the same collective area to reduce
the amount of mouse movement to go backwards, forwards, to the very
beginning, and the very end.

On Jun 8, 2007, at 10:37 AM, Allison Beckwith wrote:

> previous link, page links, next link
>
> I am wondering if this is really the best way to organize this. I
> have always thought that page links and previous/next links should be
> kept separate. Like so:
>
> page links (left aligned) previous link, next link (right
> aligned)

Cheers!

Todd Zaki Warfel
President, Design & Usability Specialist
Messagefirst | Designing Information. Beautifully.
----------------------------------
Contact Info
Voice: (215) 825-7423
Email: todd at messagefirst.com
AIM: twarfel at mac.com
Blog: http://toddwarfel.com
----------------------------------
In theory, theory and practice are the same.
In practice, they are not.

8 Jun 2007 - 4:09pm
Mark Schraad
2006

Todd's example is what I have been using (in spite of not being the
official standards method) with the one exception that I limit it to
1 2 3 - last, instead of showing 1 2 3 4 5.

On Jun 8, 2007, at 1:56 PM, Todd Zaki Warfel wrote:

> We've actually tested a number of different types of pagination
> models and have settled on the following based on time, effort, and
> satisfaction performance measurements from several rounds of testing
> on a variety of different products over the course of the past 6
> years:
>
> Default: < 1 2 3 4 5 > of (N) (where N is the last page)
>
> Once the person has moved outside the initial first five: 1 < 6 7 8 9
> 10 > (N) (where 1 is the first page and N is the last page)
>
> Once the person has reached the end of the list of pages 1 < 12 13 14
> (where 1 is the fist page and 14 is the last page and you're on 14 so
> there is no next)
>
> What we found was that the general goal and behavior was to:
> * get to the first page
> * go to the next page
> * jump to the last page
> * view ahead to see how many pages are left
>
> And we keep them all together in the same collective area to reduce
> the amount of mouse movement to go backwards, forwards, to the very
> beginning, and the very end.
>
> On Jun 8, 2007, at 10:37 AM, Allison Beckwith wrote:
>
>> previous link, page links, next link
>>
>> I am wondering if this is really the best way to organize this. I
>> have always thought that page links and previous/next links should be
>> kept separate. Like so:
>>
>> page links (left aligned) previous link, next link (right
>> aligned)
>
>
> Cheers!
>
> Todd Zaki Warfel
> President, Design & Usability Specialist
> Messagefirst | Designing Information. Beautifully.
> ----------------------------------
> Contact Info
> Voice: (215) 825-7423
> Email: todd at messagefirst.com
> AIM: twarfel at mac.com
> Blog: http://toddwarfel.com
> ----------------------------------
> In theory, theory and practice are the same.
> In practice, they are not.
>
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://listguide.ixda.org/
> List Help .................. http://listhelp.ixda.org/
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org

9 Jun 2007 - 3:31am
natekendrick
2005

I've always designed pagination like this:

< Previous Page | [ drop down menu to select page]\ | Next Page >

This was done to save horizontal space using a single drop down
instead of a large set of individual links: 1-10+ and other minor
interaction things like: the links are harder to target than a
standard drop down menu.

I haven't had a chance to user test this particular widget, but
pagination is so ubiquitous across websites, it would be interesting
to develop a single standard. A standard individual designers could
utilize across their clients/projects.

-N

On Jun 8, 2007, at 2:09 PM, Mark Schraad wrote:

> Todd's example is what I have been using (in spite of not being the
> official standards method) with the one exception that I limit it to
> 1 2 3 - last, instead of showing 1 2 3 4 5.
>
>
> On Jun 8, 2007, at 1:56 PM, Todd Zaki Warfel wrote:
>
>> We've actually tested a number of different types of pagination
>> models and have settled on the following based on time, effort, and
>> satisfaction performance measurements from several rounds of testing
>> on a variety of different products over the course of the past 6
>> years:
>>
>> Default: < 1 2 3 4 5 > of (N) (where N is the last page)
>>
>> Once the person has moved outside the initial first five: 1 < 6 7 8 9
>> 10 > (N) (where 1 is the first page and N is the last page)
>>
>> Once the person has reached the end of the list of pages 1 < 12 13 14
>> (where 1 is the fist page and 14 is the last page and you're on 14 so
>> there is no next)
>>
>> What we found was that the general goal and behavior was to:
>> * get to the first page
>> * go to the next page
>> * jump to the last page
>> * view ahead to see how many pages are left
>>
>> And we keep them all together in the same collective area to reduce
>> the amount of mouse movement to go backwards, forwards, to the very
>> beginning, and the very end.
>>
>> On Jun 8, 2007, at 10:37 AM, Allison Beckwith wrote:
>>
>>> previous link, page links, next link
>>>
>>> I am wondering if this is really the best way to organize this. I
>>> have always thought that page links and previous/next links
>>> should be
>>> kept separate. Like so:
>>>
>>> page links (left aligned) previous link, next link (right
>>> aligned)
>>
>>
>> Cheers!
>>
>> Todd Zaki Warfel
>> President, Design & Usability Specialist
>> Messagefirst | Designing Information. Beautifully.
>> ----------------------------------
>> Contact Info
>> Voice: (215) 825-7423
>> Email: todd at messagefirst.com
>> AIM: twarfel at mac.com
>> Blog: http://toddwarfel.com
>> ----------------------------------
>> In theory, theory and practice are the same.
>> In practice, they are not.
>>
>>
>> ________________________________________________________________
>> Welcome to the Interaction Design Association (IxDA)!
>> To post to this list ....... discuss at ixda.org
>> List Guidelines ............ http://listguide.ixda.org/
>> List Help .................. http://listhelp.ixda.org/
>> (Un)Subscription Options ... http://subscription-options.ixda.org/
>> Announcements List ......... http://subscribe-announce.ixda.org/
>> Questions .................. lists at ixda.org
>> Home ....................... http://ixda.org/
>> Resource Library ........... http://resources.ixda.org
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://listguide.ixda.org/
> List Help .................. http://listhelp.ixda.org/
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org

10 Jun 2007 - 11:18am
Todd Warfel
2003

One of the models we tested included a drop/select menu. The biggest
performance drawback is that it requires users to click the menu,
then select a page, release (then possibly select a go button).

Pros
-> Less space on the page

Cons
-> Requires an action to expose the options/pages you can select
-> Requires selection from a select menu. It's no unusual we see
people mis-selecting a page and going to the wrong one. This
increases frustration and error rate.
-> Requires no less than 2-3 steps to accomplish what can be done in
one with a model that has a span of pages exposed on screen (e.g. 1 2
3 4 5)

On Jun 9, 2007, at 4:31 AM, Nathan Kendrick wrote:

> This was done to save horizontal space using a single drop down
> instead of a large set of individual links: 1-10+ and other minor
> interaction things like: the links are harder to target than a
> standard drop down menu.

Cheers!

Todd Zaki Warfel
President, Design & Usability Specialist
Messagefirst | Designing Information. Beautifully.
----------------------------------
Contact Info
Voice: (215) 825-7423
Email: todd at messagefirst.com
AIM: twarfel at mac.com
Blog: http://toddwarfel.com
----------------------------------
In theory, theory and practice are the same.
In practice, they are not.

10 Jun 2007 - 2:29pm
bmeunier
2007

The best method that my group came up is:

Previous 1 - 2 3 4 5 - 57 Next

- The first page (1) and the last page (57) are always there
- On the first or last page, previous/next aren't link, but text
(previous is a text on page one, next is a text on last page)

Pros:

- Nothing disapear, always the same patterns so the user don't have
any surprise
- Very simple, less space on the page
- You've have all what you need (previous, first, last, next)

Don't forget to:

Font size minimum 16 px (or 120%)

Benoît Meunier
www.benoitmeunier.info

On 10-Jun-07, at 12:18 PM, Todd Zaki Warfel wrote:

> One of the models we tested included a drop/select menu. The biggest
> performance drawback is that it requires users to click the menu,
> then select a page, release (then possibly select a go button).
>
> Pros
> -> Less space on the page
>
> Cons
> -> Requires an action to expose the options/pages you can select
> -> Requires selection from a select menu. It's no unusual we see
> people mis-selecting a page and going to the wrong one. This
> increases frustration and error rate.
> -> Requires no less than 2-3 steps to accomplish what can be done in
> one with a model that has a span of pages exposed on screen (e.g. 1 2
> 3 4 5)
>
>
> On Jun 9, 2007, at 4:31 AM, Nathan Kendrick wrote:
>
>> This was done to save horizontal space using a single drop down
>> instead of a large set of individual links: 1-10+ and other minor
>> interaction things like: the links are harder to target than a
>> standard drop down menu.
>
>
> Cheers!
>
> Todd Zaki Warfel
> President, Design & Usability Specialist
> Messagefirst | Designing Information. Beautifully.
> ----------------------------------
> Contact Info
> Voice: (215) 825-7423
> Email: todd at messagefirst.com
> AIM: twarfel at mac.com
> Blog: http://toddwarfel.com
> ----------------------------------
> In theory, theory and practice are the same.
> In practice, they are not.
>
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://listguide.ixda.org/
> List Help .................. http://listhelp.ixda.org/
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org
>

10 Jun 2007 - 6:22pm
Kyle Cooney
2006

Generally, I prefer a method similar to what Benoit has outlined here,
althought slightly modified:

1 - 5 6 7 8 9 10 11 12 - 57 Previous || Next

I stick with maximum of 10 page links shown at any one time, with links to
the first and last pages available, as well as the 3 previous pages to the
link the user is currently on, and the next 4 links. Also, lately, I think
my preference has gone to putting the Previous and Next links together, off
a bit to the side, and having them in consistent location for users who use
them exclusively and don't want to keep repositioning their mouse.. But I
could easily see wanting to put them on either side of the 10 individual
links.

Also, for context, I generally prefer number paginations in search results
and data intensive environments. The pull down menu I think works in
articles where you're dealing with narrative chapter titles for each page,
though, even still, I tend to prefer displaying the titles outside of pull
down for the reasons outlined by Mr. Warfel.

Kyle

On 6/10/07, Benoît Meunier <ecrire at benoitmeunier.info> wrote:
>
> The best method that my group came up is:
>
> Previous 1 - 2 3 4 5 - 57 Next
>
> - The first page (1) and the last page (57) are always there
> - On the first or last page, previous/next aren't link, but text
> (previous is a text on page one, next is a text on last page)
>
> Pros:
>
> - Nothing disapear, always the same patterns so the user don't have
> any surprise
> - Very simple, less space on the page
> - You've have all what you need (previous, first, last, next)
>
> Don't forget to:
>
> Font size minimum 16 px (or 120%)
>
> Benoît Meunier
> www.benoitmeunier.info
>
>
>
> On 10-Jun-07, at 12:18 PM, Todd Zaki Warfel wrote:
>
> > One of the models we tested included a drop/select menu. The biggest
> > performance drawback is that it requires users to click the menu,
> > then select a page, release (then possibly select a go button).
> >
> > Pros
> > -> Less space on the page
> >
> > Cons
> > -> Requires an action to expose the options/pages you can select
> > -> Requires selection from a select menu. It's no unusual we see
> > people mis-selecting a page and going to the wrong one. This
> > increases frustration and error rate.
> > -> Requires no less than 2-3 steps to accomplish what can be done in
> > one with a model that has a span of pages exposed on screen (e.g. 1 2
> > 3 4 5)
> >
> >
> > On Jun 9, 2007, at 4:31 AM, Nathan Kendrick wrote:
> >
> >> This was done to save horizontal space using a single drop down
> >> instead of a large set of individual links: 1-10+ and other minor
> >> interaction things like: the links are harder to target than a
> >> standard drop down menu.
> >
> >
> > Cheers!
> >
> > Todd Zaki Warfel
> > President, Design & Usability Specialist
> > Messagefirst | Designing Information. Beautifully.
> > ----------------------------------
> > Contact Info
> > Voice: (215) 825-7423
> > Email: todd at messagefirst.com
> > AIM: twarfel at mac.com
> > Blog: http://toddwarfel.com
> > ----------------------------------
> > In theory, theory and practice are the same.
> > In practice, they are not.
> >
> >
> > ________________________________________________________________
> > Welcome to the Interaction Design Association (IxDA)!
> > To post to this list ....... discuss at ixda.org
> > List Guidelines ............ http://listguide.ixda.org/
> > List Help .................. http://listhelp.ixda.org/
> > (Un)Subscription Options ... http://subscription-options.ixda.org/
> > Announcements List ......... http://subscribe-announce.ixda.org/
> > Questions .................. lists at ixda.org
> > Home ....................... http://ixda.org/
> > Resource Library ........... http://resources.ixda.org
> >
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://listguide.ixda.org/
> List Help .................. http://listhelp.ixda.org/
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org
>

--
Of Interest: http://del.icio.us/kccdc
ph: 202.270.8214
2505 13th Street NW Apt #411
Washington, DC
20009

10 Jun 2007 - 6:42pm
dmitryn
2004

Todd,

Out of curiousity, in your testing, did you measure the proportion of
users who relied on the Previous/Next buttons vs. the numbered page
links?

Thanks,

Dmitry

On 6/10/07, Todd Zaki Warfel <lists at toddwarfel.com> wrote:
> One of the models we tested included a drop/select menu. The biggest
> performance drawback is that it requires users to click the menu,
> then select a page, release (then possibly select a go button).
>
> Pros
> -> Less space on the page
>
> Cons
> -> Requires an action to expose the options/pages you can select
> -> Requires selection from a select menu. It's no unusual we see
> people mis-selecting a page and going to the wrong one. This
> increases frustration and error rate.
> -> Requires no less than 2-3 steps to accomplish what can be done in
> one with a model that has a span of pages exposed on screen (e.g. 1 2
> 3 4 5)
>
>
> On Jun 9, 2007, at 4:31 AM, Nathan Kendrick wrote:
>
> > This was done to save horizontal space using a single drop down
> > instead of a large set of individual links: 1-10+ and other minor
> > interaction things like: the links are harder to target than a
> > standard drop down menu.
>
>
> Cheers!
>
> Todd Zaki Warfel
> President, Design & Usability Specialist
> Messagefirst | Designing Information. Beautifully.
> ----------------------------------
> Contact Info
> Voice: (215) 825-7423
> Email: todd at messagefirst.com
> AIM: twarfel at mac.com
> Blog: http://toddwarfel.com
> ----------------------------------
> In theory, theory and practice are the same.
> In practice, they are not.
>
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://listguide.ixda.org/
> List Help .................. http://listhelp.ixda.org/
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org
>

10 Jun 2007 - 6:44pm
Todd Warfel
2003

The biggest drawback here is that from a mouse targeting perspective
(fitz's law), you're requiring someone to move halfway across the
screen to move through the pages by Next/Prev or page number. Other
than that...

On Jun 10, 2007, at 7:22 PM, Kyle Cooney wrote:

> 1 - 5 6 7 8 9 10 11 12 - 57 Previous ||
> Next

Cheers!

Todd Zaki Warfel
President, Design & Usability Specialist
Messagefirst | Designing Information. Beautifully.
----------------------------------
Contact Info
Voice: (215) 825-7423
Email: todd at messagefirst.com
AIM: twarfel at mac.com
Blog: http://toddwarfel.com
----------------------------------
In theory, theory and practice are the same.
In practice, they are not.

10 Jun 2007 - 6:46pm
Todd Warfel
2003

Depends on the context, but in general, most use prev/next combined
with the browser back button.

Use of numbers tended to be when they were moving through large
numbers of search results or ecommerce sites and wanted to get "back
to" something that was on a previous screen and there wasn't a
"recently viewed" module available. There are a few other cases, but
for the most part, the majority of navigation is w/the use of
supplied prev/next and the browser's back button.

On Jun 10, 2007, at 7:42 PM, Dmitry Nekrasovski wrote:

> Todd,
>
> Out of curiousity, in your testing, did you measure the proportion of
> users who relied on the Previous/Next buttons vs. the numbered page
> links?
>
> Thanks,
>
> Dmitry

Cheers!

Todd Zaki Warfel
President, Design & Usability Specialist
Messagefirst | Designing Information. Beautifully.
----------------------------------
Contact Info
Voice: (215) 825-7423
Email: todd at messagefirst.com
AIM: twarfel at mac.com
Blog: http://toddwarfel.com
----------------------------------
In theory, theory and practice are the same.
In practice, they are not.

11 Jun 2007 - 12:07am
AlokJain
2006

One of the models that we had briefly experimented with was using a
combination of page numbers and a "slider"

[1] 2 3 4 5 ----/\--------
12

The idea was that after initial set of pages, people work with less
precision, which is in relation to complete number of pages. For
instance, people might try to look "toward the end of the results",
"Jump a few pages". By showing a slider it provides the context of
total number of pages and provides that less precise functioning. As
the slider is moved the page number is displayed as well (shown as 12
in above e.g.)

I could not user test before I left the company, so it is not been
validated yet.

Best Regards
Alok Jain
----------------------------------
http://www.iPrincipia.com

On Jun 10, 2007, at 9:48 PM, Todd Zaki Warfel wrote:

> One of the models we tested included a drop/select menu. The biggest
> performance drawback is that it requires users to click the menu,
> then select a page, release (then possibly select a go button).

11 Jun 2007 - 9:40am
Valerie Gomez d...
2006

This is very interesting. Thank you all for sharing.

One question: Has anyone done any testing around the size of the pagination
control? It has always seemed to me that pagination is something users will
use quite a bit when going through pages of information and yet we make the
control to navigate through these pages fairly small to click on.

Thoughts?

11 Jun 2007 - 9:57am
bmeunier
2007

For governmentbids.com ex.: (http://www.governmentbids.com/cgi/en/browse.category.list/1175-1076.htm) we've done tree tests:

12px
14px
16px

The users have click more with a font size of 14px, less with 16px under 10 pages results.
Over 10 pages, the best font-size is 12px (110%).

Benoit Meunier
+1 514.962.2773

www.benoitmeunier.info

> -------- Original Message --------
> Subject: Re: [IxDA Discuss] Pagination
> From: "Valerie Gomez de la Torre" <valeriegomez at gmail.com>
> Date: Mon, June 11, 2007 7:40 am
> To: "Alok Jain" <alok.ajain1 at gmail.com>
> Cc: IxDA Discuss <discuss at ixda.org>
>
> This is very interesting. Thank you all for sharing.
>
> One question: Has anyone done any testing around the size of the
> pagination
> control? It has always seemed to me that pagination is something users
> will
> use quite a bit when going through pages of information and yet we
> make the
> control to navigate through these pages fairly small to click on.
>
> Thoughts?
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://listguide.ixda.org/
> List Help .................. http://listhelp.ixda.org/
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org

11 Jun 2007 - 11:44am
Todd Warfel
2003

That could be an interesting one to test.

On Jun 11, 2007, at 1:07 AM, Alok Jain wrote:

> One of the models that we had briefly experimented with was using a
> combination of page numbers and a "slider"
>
> [1] 2 3 4 5 ----/\--------
> 12

Cheers!

Todd Zaki Warfel
President, Design & Usability Specialist
Messagefirst | Designing Information. Beautifully.
----------------------------------
Contact Info
Voice: (215) 825-7423
Email: todd at messagefirst.com
AIM: twarfel at mac.com
Blog: http://toddwarfel.com
----------------------------------
In theory, theory and practice are the same.
In practice, they are not.

11 Jun 2007 - 12:28pm
Fredrik Matheson
2005

I've tried out a number of paginators and so far my favorite, for lists that
are long and hard to filte looks like this:
| Previous | 1 … 11 12 13 14 [15] 16 17 18 19 … 107 pages | Next | Jump to page [ ]

For true previous/next navigational efficacy, "B" (back) and "N" (next)
keyboard shortcuts are pretty much unbeatable. No looking for targets, no
fumbling with the mouse. Fitt's law would suggest putting Previous and Next
together, but I've found it messes up the logic of the paginator a little.

One huge problem that the "disjointed linear paginator" has is jumping to
another page. Drop-downs are one alternative, as mentioned, but I think a
text field works even better. Just type in what page you want and off you
go.

The only problem with ALL of this is that we're using a broken model to work
around technical constraints. After all, the pages are merely rendered as
discrete object, but in reality there's a very long stream of stuff.

In a book pages are great for finding where you left off. On web sites,
they're pretty useless for this because the content will frequently be added
to and the page content reshuffled, so now something else will be on the
page. Pagination, then, only offers chunking options ("I'll read just one
more page") and gives a loose estimate of how much stuff is ahead of you (a
good thing)

Now, compare that to Google Reader, where new items are merely loaded into
the viewable area as you move downwards. There's no flashing of white
between pages and no needless refreshing of page frameworks.

In a current project, I've implemented a slider that rolls a dataset (a long
table) through a viewing frame. The slider moves horizontally while the
content moves vertically and the slider has an indicator that shows where
you are in the list. For our particlular problem, it has worked very well.
There's no pagination, no "where should I go?" cognitive load, it takes up
very little space on the screen (space is at a premium in this context) and
the users grasp it at once. Best of all, it reduces the amount of time that
users need to interact with the system to get what they need.

Has anyone here tried Google Reader-type content sliding on a larger scale?
What are your experiences?

- Fredrik

11 Jun 2007 - 2:34pm
cfmdesigns
2004

I agree that this looks both good and useful. I used to work for an eBook company.

When a user is on page N, the things he most wants to do are:

* Read the next page
* Skip back to the previous page, to re-establish context
* Skip back or forward a few pages, to crosscheck stuff in the narrative
* Know where the next section (chapter) break is
* Go to the index or table of contents
* Know how many pages are in the whole work ("Am I almost done, or is this War & Peace?")
* Randomly skip ahead, the equivalent of leafing through a book to see what catches the eye
* Set a bookmark to come back to a specific place

It looks like your solution, or variations on it, satisfy many of these needs. (And some, of course, don't apply in some situations. For example, Google Search results: unless you know that the link you want is on page 5, no one is going to skip ahead to a later page in the results, lest they miss the link they need. The slider solution, and even almost anything but previous/next/how many pages, are not warranted there, I think.)

-- Jim Drew
Seattle, WA

-----Original Message-----
>From: Alok Jain <alok.ajain1 at gmail.com>
>
>One of the models that we had briefly experimented with was using a
>combination of page numbers and a "slider"
>
> [1] 2 3 4 5 ----/\--------
> 12
>
>The idea was that after initial set of pages, people work with less

11 Jun 2007 - 3:40pm
bmeunier
2007

Don't forget about accessibility:

- The community have give us some default access key to use. Be sure
to not use them! (1: Return to Home page, 3: Site Map, 9: Contact,
etc.)<http://www.mediclub.ca/en/contact/>
- Those slides are difficult to use for those who have some physical
problems.

Benoit Meunier

On 6/11/07, Jim Drew <cfmdesigns at earthlink.net> wrote:
>
> I agree that this looks both good and useful. I used to work for an eBook
> company.
>
> When a user is on page N, the things he most wants to do are:
>
> * Read the next page
> * Skip back to the previous page, to re-establish context
> * Skip back or forward a few pages, to crosscheck stuff in the narrative
> * Know where the next section (chapter) break is
> * Go to the index or table of contents
> * Know how many pages are in the whole work ("Am I almost done, or is this
> War & Peace?")
> * Randomly skip ahead, the equivalent of leafing through a book to see
> what catches the eye
> * Set a bookmark to come back to a specific place
>
> It looks like your solution, or variations on it, satisfy many of these
> needs. (And some, of course, don't apply in some situations. For example,
> Google Search results: unless you know that the link you want is on page 5,
> no one is going to skip ahead to a later page in the results, lest they miss
> the link they need. The slider solution, and even almost anything but
> previous/next/how many pages, are not warranted there, I think.)
>
> -- Jim Drew
> Seattle, WA
>
>
> -----Original Message-----
> >From: Alok Jain <alok.ajain1 at gmail.com>
> >
> >One of the models that we had briefly experimented with was using a
> >combination of page numbers and a "slider"
> >
> > [1] 2 3 4 5 ----/\--------
> > 12
> >
> >The idea was that after initial set of pages, people work with less
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://listguide.ixda.org/
> List Help .................. http://listhelp.ixda.org/
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org
>
>

--
Benoît Meunier
+1 514-574-3142

www.benoitmeunier.info

15 Jun 2007 - 11:52am
Valerie Gomez d...
2006

Hi Todd - Can you tell me where this pagination was being used?

To go through search results or a queue of data or...

Thank you,
Valerie

Default: < 1 2 3 4 5 > of (N) (where N is the last page)
>
> Once the person has moved outside the initial first five: 1 < 6 7 8 9
> 10 > (N) (where 1 is the first page and N is the last page)
>
> Once the person has reached the end of the list of pages 1 < 12 13 14
> (where 1 is the fist page and 14 is the last page and you're on 14 so
> there is no next)
>
> What we found was that the general goal and behavior was to:
> * get to the first page
> * go to the next page
> * jump to the last page
> * view ahead to see how many pages are left
>
>
>

18 Jun 2007 - 8:50am
Todd Warfel
2003

Primarily:
* Search results
* Call logs for VoIP system
* Bank statements

There are other areas we've used this in, but these are some specific
examples.

On Jun 15, 2007, at 12:52 PM, Valerie Gomez de la Torre wrote:

> Hi Todd - Can you tell me where this pagination was being used?
>
> To go through search results or a queue of data or...
>
> Thank you,
> Valerie

Cheers!

Todd Zaki Warfel
President, Design & Usability Specialist
Messagefirst | Designing Information. Beautifully.
----------------------------------
Contact Info
Voice: (215) 825-7423
Email: todd at messagefirst.com
AIM: twarfel at mac.com
Blog: http://toddwarfel.com
----------------------------------
In theory, theory and practice are the same.
In practice, they are not.

18 Jun 2007 - 9:25am
Chris Pallé
2007

Have you (or anyone else) had any experience with integrating this
with a "View N Items" per page control?

chris.pallé, user experience
--------------------------------------------------------
blueflameinteractive
732.513.3570
chris.palle at blueflameinteractive.com
http://blueflameinteractive.com
http://www.linkedin.com/in/chrispalle

On Jun 8, 2007, at 1:56 PM, Todd Zaki Warfel wrote:

>
> Default: < 1 2 3 4 5 > of (N) (where N is the last page)
>
> Once the person has moved outside the initial first five: 1 < 6 7 8 9
> 10 > (N) (where 1 is the first page and N is the last page)
>
> Once the person has reached the end of the list of pages 1 < 12 13 14
> (where 1 is the fist page and 14 is the last page and you're on 14 so
> there is no next)
>
> What we found was that the general goal and behavior was to:
> * get to the first page
> * go to the next page
> * jump to the last page
> * view ahead to see how many pages are left
>
> And we keep them all together in the same collective area to reduce
> the amount of mouse movement to go backwards, forwards, to the very
> beginning, and the very end.

18 Jun 2007 - 4:02pm
Tanya Rabourn
2004

On 6/11/07, Jim Drew <cfmdesigns at earthlink.net> wrote:
> I agree that this looks both good and useful. I used to work for an eBook company.
>
> When a user is on page N, the things he most wants to do are:
>

What about cite? When constructing eBooks (on projects with a
university press) that was one of the main problems we encountered
when trying to break out of a strict page paradigm. I believe we went
with a numbered paragraph approach that a few eBook companies use.
That seems rather unsatisfactory too though.

> * Set a bookmark to come back to a specific place

Did this also solve the citation problem or was there no visible marker?

-Tanya

18 Jun 2007 - 5:53pm
cfmdesigns
2004

>From: Tanya Rabourn <tanya at pixelcharmer.com>
>
>On 6/11/07, Jim Drew <cfmdesigns at earthlink.net> wrote:
>> I agree that this looks both good and useful. I used to work for an eBook company.
>>
>> When a user is on page N, the things he most wants to do are:
>
>What about cite? When constructing eBooks (on projects with a
>university press) that was one of the main problems we encountered
>when trying to break out of a strict page paradigm. I believe we went
>with a numbered paragraph approach that a few eBook companies use.
>That seems rather unsatisfactory too though.

I think that citation falls outside the realm of what a user wants to do next regarding a pagination control.

It's certainly an interesting and problematic issue, of course. Citation is reasonably easy in a fixed-page situation (PDF), doable in a situation where there are a known set of variants (Microsoft Reader, SoftBook Reader, RocketBook, etc.), and difficult where the text varies base on user settings -- font size, browser width, etc. eBook lists in 1999-2000 had some major threads on the subject, some of them centered around the same issue in printed matter, such as when students in a class may be using different published versions of the same work, so you can't say "Turn to page 119" and have everyone on, er, the same page.

>> * Set a bookmark to come back to a specific place
>
>Did this also solve the citation problem or was there no visible marker?

It can be used to solve (or at least minimize) the issue. User-insertable named markers, or at least a set of user-creatable markers which can be tied to immutable characteristics of the substratum of the text (such as unique IDs in the <DIV> tags for an HTML presentation), seemed to be the best answer.

The recognition was that "section 5, subsection 12, paragraph 3, line 4" is *not* the goal; the goal is "The horror, the horror" in Joseph Conrad's "The Heart of Darkness". But many people tie themselves so tightly to a format that they cannot divorce themselves from seeing the format as the goal itself, and thus that there are many potential ways of getting to/referring to the same passage.

And thus, when dealing with an electronic text, pagination is meaningless, and artificially imposed format on the stream. (We see that in things like bank statements and webmail and such: you don't want to 3rd item on the 5th page, you want item #83, paginated however works for you.) Thus citations in electronic works need to be user-friendly (textual) and connected to parts of the text which retain their significance regardless of user preferences.

-- Jim

20 Jun 2007 - 2:15pm
Vishal Subraman...
2005

On a different note, I've noticed pagination only being used at the
bottom of search results in most major websites (Google, Yahoo
Shopping etc). Don't you think it should be both on top and bottom of
the search results?

--
Vishal Iyer
http://www.vishaliyer.com

20 Jun 2007 - 2:56pm
Alexander Baxevanis
2007

I can't see any good reason for putting pagination on top for search
results: don't you need to at least take a look at the first page of
results (who are supposed to be the most "relevant") before deciding
to go further?

Cheers,

Alex

On 6/20/07, Vishal Iyer <vishaliyer1 at gmail.com> wrote:
> On a different note, I've noticed pagination only being used at the
> bottom of search results in most major websites (Google, Yahoo
> Shopping etc). Don't you think it should be both on top and bottom of
> the search results?
>
> --
> Vishal Iyer
> http://www.vishaliyer.com
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://listguide.ixda.org/
> List Help .................. http://listhelp.ixda.org/
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org
>

20 Jun 2007 - 3:04pm
Vishal Subraman...
2005

You dont have to remain at the bottom of the page after scanning it
once to want to go to another page. Why make the user scroll down to
the bottom if s/he is closer to the top?

> I can't see any good reason for putting pagination on top for search
> results: don't you need to at least take a look at the first page of
> results (who are supposed to be the most "relevant") before deciding
> to go further?

--
-Vishal
http://www.vishaliyer.com

20 Jun 2007 - 3:05pm
Jack L. Moffett
2005

On Jun 20, 2007, at 3:56 PM, Alexander Baxevanis wrote:

> I can't see any good reason for putting pagination on top for search
> results: don't you need to at least take a look at the first page of
> results (who are supposed to be the most "relevant") before deciding
> to go further?

But when you are navigating back, say to find a result you passed up,
and the navigation at the bottom is below the fold, navigation at the
top is very helpful. I've noticed that Amazon puts navigation at the
top and bottom of Wish List pages, but only at the bottom of search
results.

Jack

Jack L. Moffett
Interaction Designer
inmedius
412.459.0310 x219
http://www.inmedius.com

Form follows function -
that has been misunderstood.
Form and function should be one,
joined in a spiritual union.

- Frank Lloyd Wright

20 Jun 2007 - 3:06pm
Oleh Kovalchuke
2006

Two reasons:
1. Repeated search with relevant result displayed on the third ("or was it
fourth?") page.
2. To pace (estimate) change in the relevance of search results.

--
Oleh Kovalchuke
Interaction Design is the Design of Time
http://www.tangospring.com/IxDtopicWhatIsInteractionDesign.htm

On 6/20/07, Alexander Baxevanis <alex.baxevanis at gmail.com> wrote:

> I can't see any good reason for putting pagination on top for search
> results: don't you need to at least take a look at the first page of
> results (who are supposed to be the most "relevant") before deciding
> to go further?
>
> Cheers,
>
> Alex
>
> On 6/20/07, Vishal Iyer <vishaliyer1 at gmail.com> wrote:
> > On a different note, I've noticed pagination only being used at the
> > bottom of search results in most major websites (Google, Yahoo
> > Shopping etc). Don't you think it should be both on top and bottom of
> > the search results?
> >
> > --
> > Vishal Iyer
> > http://www.vishaliyer.com
>
>

20 Jun 2007 - 4:31pm
cfmdesigns
2004

I don't know, *do* you need to look before moving on? In every possible case?

Do you need to look if you've already looked but you've come back to the page? Do you need to look if you know it's almost certainly not on that page (such as when you know there are 1000 entries, 20 per page, these start with "A", and you need something starting with "R")? Do you need to look if you've scanned the page and come back to the top for whatever reason?

In the end, if you use the same controls in both places, the hit to show them in two spots is minimal, so reasons to not do it relating to cost shouldn't apply. And if it will serve some users some times, it may be worthwhile.

On the other hand, if your information is truly geared such that each page must be scanned completely, top to bottom, and no one would need to go up again (I find this doubtful), or if your information is limited such that the bottom can never be "below the fold" (fixed window and font sizes, etc.), then only using one set of controls may make sense.

(To be sure, there's no need to put the controls in 18 places on the screen, just because a user's mouse might be near that space.)

-- Jim

>From: Alexander Baxevanis <alex.baxevanis at gmail.com>
>
>I can't see any good reason for putting pagination on top for search
>results: don't you need to at least take a look at the first page of
>results (who are supposed to be the most "relevant") before deciding
>to go further?
>
>
>On 6/20/07, Vishal Iyer <vishaliyer1 at gmail.com> wrote:
>> On a different note, I've noticed pagination only being used at the
>> bottom of search results in most major websites (Google, Yahoo
>> Shopping etc). Don't you think it should be both on top and bottom of
>> the search results?

22 Jun 2007 - 9:18am
Chris Pallé
2007

Not sure if anyone else posted this, but I thought it had some good
observations:

http://kurafire.net/log/archive/2007/06/22/pagination-101

chris.pallé, user experience
--------------------------------------------------------
blueflameinteractive
732.513.3570
chris.palle at blueflameinteractive.com
http://blueflameinteractive.com
http://www.linkedin.com/in/chrispalle

On Jun 20, 2007, at 5:31 PM, Jim Drew wrote:

> I don't know, *do* you need to look before moving on? In every
> possible case?
>
> Do you need to look if you've already looked but you've come back
> to the page? Do you need to look if you know it's almost certainly
> not on that page (such as when you know there are 1000 entries, 20
> per page, these start with "A", and you need something starting
> with "R")? Do you need to look if you've scanned the page and come
> back to the top for whatever reason?
>
> In the end, if you use the same controls in both places, the hit to
> show them in two spots is minimal, so reasons to not do it relating
> to cost shouldn't apply. And if it will serve some users some
> times, it may be worthwhile.
>
> On the other hand, if your information is truly geared such that
> each page must be scanned completely, top to bottom, and no one
> would need to go up again (I find this doubtful), or if your
> information is limited such that the bottom can never be "below the
> fold" (fixed window and font sizes, etc.), then only using one set
> of controls may make sense.
>
> (To be sure, there's no need to put the controls in 18 places on
> the screen, just because a user's mouse might be near that space.)
>
> -- Jim
>
>
>> From: Alexander Baxevanis <alex.baxevanis at gmail.com>
>>
>> I can't see any good reason for putting pagination on top for search
>> results: don't you need to at least take a look at the first page of
>> results (who are supposed to be the most "relevant") before deciding
>> to go further?
>>
>>
>> On 6/20/07, Vishal Iyer <vishaliyer1 at gmail.com> wrote:
>>> On a different note, I've noticed pagination only being used at the
>>> bottom of search results in most major websites (Google, Yahoo
>>> Shopping etc). Don't you think it should be both on top and
>>> bottom of
>>> the search results?
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://listguide.ixda.org/
> List Help .................. http://listhelp.ixda.org/
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org

22 Jun 2007 - 9:51am
Dan Brown
2004

Far be it for me to start a controversy, but pagination seems more
necessary evil than useful interaction element. While I appreciate the
depth of thought that goes into these analyses, but I'd be more
interested in seeing examples of *meaningful* navigation for a a group
of items of a particular type.

A "page" in a list of results is a more-or-less artificial grouping of
items. Those items at the top of the list -- presumably on the first
page -- are judged by the underlying alogrithm to be the Most
Important.

But, who are we kidding?

While the examples provided in this thread have been exemplary, they
all seem to break down when facing large numbers of items because
"page 27" just isn't meaningful to people when their looking for
something in a list.

For a previous job, I looked into alternate methods for navigating
long lists of items. Flickr (at the time) provided an interesting
paradigm in their Organizr -- the timeline, which displayed a bar
chart of photos uploaded over time. You could move sliders around to
zero-in on a particular time. Long list of items, meaningful
one-dimensional navigation. I extended the idea in my particular case
to allow users to navigate by longitude. (This may seem a little
crazy, but it was actually a meaningful dimension to our target
audience.)

Thought exercise: take a group of items you're dealing with and
identify a meaningful single dimension along which users might
navigate... Selling clothing online? Maybe users can walk along a
spectrum of colors (or use a slider control to zero-in on preferred
colors)...

So, will pagination become increasingly irrelevant, or am I just fooling myself?

-- Dan

On 6/22/07, Chris Pallé <chris.palle at blueflameinteractive.com> wrote:
> Not sure if anyone else posted this, but I thought it had some good
> observations:
>
> http://kurafire.net/log/archive/2007/06/22/pagination-101
>

--
} work: eightshapes.com
} book: communicatingdesign.com
} blog: greenonions.com
} talk: +1 (301) 801-4850

22 Jun 2007 - 10:11am
Mark Schraad
2006

Dan,

As I read your post, I find some assumptions that apply only to some uses of pagination. In the case of long text articles... you can assume a journalistic structure that provides for more important content at the top (or first page) of the article, but that is not always the case.

In many filter/listing functions, the ordering is by distance/location, price, size, or name/alpha. In this case page 27 may in fact be the most important and while a sorting mechanism helps, it is not always useful. There is an strong relationship between 'filter again' and 'pagination' that I believe is the sweet spot of managing this content (at both the user and UI designer). The context of the task, as well as the use conditions (are they in a hurry, do they have all day) and goals will drive the ideal thresholds here.

For search purposes... I think you have a decent point, but I continually come across topics that need to be searched where there is considerable overlap in terminology, but many of the results are not relevant. Filtering those out with boolean functions is not always effective.

I may be short sighted in anticipating future search/filter technologies, but as long as there is a finite display size, I believe that pagination will be necessary.

Mark

On Friday, June 22, 2007, at 10:52AM, "Dan Brown" <brownorama at gmail.com> wrote:
>Far be it for me to start a controversy, but pagination seems more
>necessary evil than useful interaction element. While I appreciate the
>depth of thought that goes into these analyses, but I'd be more
>interested in seeing examples of *meaningful* navigation for a a group
>of items of a particular type.
>
>A "page" in a list of results is a more-or-less artificial grouping of
>items. Those items at the top of the list -- presumably on the first
>page -- are judged by the underlying alogrithm to be the Most
>Important.
>
>But, who are we kidding?
>
>While the examples provided in this thread have been exemplary, they
>all seem to break down when facing large numbers of items because
>"page 27" just isn't meaningful to people when their looking for
>something in a list.
>
>For a previous job, I looked into alternate methods for navigating
>long lists of items. Flickr (at the time) provided an interesting
>paradigm in their Organizr -- the timeline, which displayed a bar
>chart of photos uploaded over time. You could move sliders around to
>zero-in on a particular time. Long list of items, meaningful
>one-dimensional navigation. I extended the idea in my particular case
>to allow users to navigate by longitude. (This may seem a little
>crazy, but it was actually a meaningful dimension to our target
>audience.)
>
>Thought exercise: take a group of items you're dealing with and
>identify a meaningful single dimension along which users might
>navigate... Selling clothing online? Maybe users can walk along a
>spectrum of colors (or use a slider control to zero-in on preferred
>colors)...
>
>So, will pagination become increasingly irrelevant, or am I just fooling myself?
>
>-- Dan
>
>
>On 6/22/07, Chris Pallé <chris.palle at blueflameinteractive.com> wrote:
>> Not sure if anyone else posted this, but I thought it had some good
>> observations:
>>
>> http://kurafire.net/log/archive/2007/06/22/pagination-101

22 Jun 2007 - 2:18pm
cfmdesigns
2004

>From: Dan Brown <brownorama at gmail.com>
>
>Far be it for me to start a controversy, but pagination seems more
>necessary evil than useful interaction element. While I appreciate the
>depth of thought that goes into these analyses, but I'd be more
>interested in seeing examples of *meaningful* navigation for a a group
>of items of a particular type.
>
>A "page" in a list of results is a more-or-less artificial grouping of
>items.

I would use the term "convenient" rather than "artificial". What pagination is, is "chunking" -- providing a stream of content in a manner that human beings can process, a chunk at a time.

That said, there is also a tendency to force pagination on streams of information where it doesn't belong or where it is not particularly useful. Inertia plays a big role in many of our UX follies.

>Those items at the top of the list -- presumably on the first
>page -- are judged by the underlying alogrithm to be the Most Important.

Or First or Easiest or whatever. There may be no intent behind what is listed; human being try to impose a narrative on whatever they see, however.

>While the examples provided in this thread have been exemplary, they
>all seem to break down when facing large numbers of items because
>"page 27" just isn't meaningful to people when their looking for
>something in a list.

I agree. "Page 27" is only useful if you know what is on/at page 27, in which case what is useful is "the marker for the stuff I want". "Page 27" is the name for that marker in the current chunking style.

The useful chunks are First, Early, Recent, Current, Next, Late, Last, and What I Want. Name them however you like.

>For a previous job, I looked into alternate methods for navigating
>long lists of items. Flickr (at the time) provided an interesting
>paradigm in their Organizr -- the timeline, which displayed a bar
>chart of photos uploaded over time. You could move sliders around to
>zero-in on a particular time.

One of the side angles on this is that chunking becomes dynamic rather than arbitrary: the chunk becomes what the use sees at the moment, based on scroll and window width.

That's what the typical problem with pagination of digital content streams is: by adhering to the printed model, the content provider takes control away from the user and says "I know best". Which inevitably isn't true, because what you oeally need to see is the last two items on tis "page" and the first one on the next, at the same time. We accept that limitation in the physical world, but then we turn and force the limitation onto the digital one and pretend it's a Good Thing.

>So, will pagination become increasingly irrelevant, or am I just fooling myself?

I think you're fooling yourself if you think that the increasing irrelevancy will come very quickly. So long as we're working on relatively fixed-size screens, and using software to create documents aimed to be printed on fixed-size sheets of paper, people are going to be mapping the physical world on the digital one and demanding the latter conform.

As with so many things, it's going to be the younger generation, in concert with forward-looking and experimenting companies who recognize the limits and try to break them, which will really change the world. They are the ones who are going to have so many options that rigid pagination will become old school and retro and thus unuseful.

-- Jim

22 Jun 2007 - 4:20pm
Christopher Fahey
2005

> What pagination is, is "chunking" -- providing a stream of
> content in a manner that human beings can process, a chunk
> at a time.
>
> That said, there is also a tendency to force pagination on
> streams of information where it doesn't belong or where it
> is not particularly useful. Inertia plays a big role in
> many of our UX follies.

Hell yeah. Pagination is a throwback. The most efficient solution I've
found is the continuously-loading-page approach, or what I call the
"bottomless page", where the "second page" of results loads below the
first page (using Ajax), then the "third page" loads below the second,
etc... all happening continuously as you scroll down the page. There is
no pagination to speak of.

Scroll down to the "bottom" of the page and keep an eye on that
scrollbar:
http://humanized.com/reader/

Of course, it's a risky new UI and it complicates ad models that require
lots of clickin (of course, a new ad can load up every thousand pixels
down the page). But that's inertia for you!

-Cf

Christopher Fahey
____________________________
Behavior
http://www.behaviordesign.com
212.532.4002 x203
646.338.4002 mobile

23 Jun 2007 - 8:02am
Todd Warfel
2003

That and there's no telling where the end is...

On Jun 22, 2007, at 5:20 PM, Christopher Fahey wrote:

> Of course, it's a risky new UI and it complicates ad models that
> require
> lots of clickin (of course, a new ad can load up every thousand pixels
> down the page). But that's inertia for you!

Cheers!

Todd Zaki Warfel
President, Design & Usability Specialist
Messagefirst | Designing Information. Beautifully.
----------------------------------
Contact Info
Voice: (215) 825-7423
Email: todd at messagefirst.com
AIM: twarfel at mac.com
Blog: http://toddwarfel.com
----------------------------------
In theory, theory and practice are the same.
In practice, they are not.

Syndicate content Get the feed